INDEX

Explanations

references to podcasts

oai_token-act-pair · gpt-3.5-turbo

mentions of podcasts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

metics

-0.73

Downloadha

-0.69

axy

-0.65

 deed

-0.65

basketball

-0.61

mbol

-0.61

pes

-0.61

ju

-0.60

hog

-0.60

 contracting

-0.60

POSITIVE LOGITS

 listeners

1.05

 listener

0.92

odcast

0.90

estern

0.87

Episode

0.87

CAST

0.87

 Patreon

0.86

 episode

0.84

 Episode

0.82

 transcript

0.82

Activations Density 0.035%

references to podcasts

mentions of podcasts

No Comments

No Known Activations